Questions tagged [forecasting]

Prediction of the future events. It is a special case of [prediction], in the context of [time-series].

Filter by
Sorted by
Tagged with
0
votes
0answers
12 views

One year of daily data - Holt Winters

I am doing basically my first forecasting model in R and I have some questions. I used Airbnb data available from this Kaggle project: https://www.kaggle.com/airbnb/seattle I had to use statistic ...
0
votes
1answer
22 views

What kind of data suits ARIMA forecasting?

I understand that an ARMA series is of the form $y_t = \mu + \phi_{1} y_{t-1} +…+ \phi_{p} y_{t-p} - \theta_{1} e_{t-1} -…- \theta_{q} e_{t-q}$ Can any form of quantitative variable ( price, ...
0
votes
0answers
11 views

SARIMA Forecast Longterm Downsides

While doing several SARIMA Forecasts I do not understand why SARIMA Forecasting is only considered to be short term forecasts and other methods like regression are longterm? What are the downsides of ...
0
votes
0answers
9 views

How can i apply Random Forest regression in time series data [duplicate]

I have daily water level data of 1990 to 2010 with precipitation,solar,temperature humidity and wind data.I want to apply random forest method in this time series data for estimating water level.But ...
0
votes
0answers
11 views

Out of Sample Forecast with Regression Formula [closed]

How do I create an out of sample forecast based on the obtained regression params. Input are the explantory variables for 10 years out of sample. Thank you. Max
1
vote
0answers
25 views

Is there a name for forecasting present and past quantities based on present data?

I'm not talking about backtesting, where you simulate forecast results based on past data only. The situation I'm talking about is when a metric about the present or past hasn't been completely ...
0
votes
2answers
34 views

How to predict future of time series 1 with time series 2 with AR(I)MA?

Hey guys i am new in this forum. I am also new into programming with R or Stata(and programming in total, but i really would like to learn it). Currently I am writing a thesis about whether its ...
1
vote
0answers
28 views

How do i forecast a time series is about to cross a certain threshold

I have data coming in 10 seconds apart which is the temperature of a given room. It has a seasonality of 6 hours as it has 2 AC switching back and forth. Sometimes the AC in this room fails and thus ...
0
votes
0answers
11 views

Pmdarima Weekday only vs Weekend Only vs Whole Week Forecasting

I want to implement Pmdarima auto arima module in my daily forecasting process(I use Python and i don't use R). Regarding to different needs sometimes i need to forecast only weekdays or only weekends ...
0
votes
0answers
12 views

Forecasting in R with future data proved to be correlated

I have been doing some work regarding differences between Australian and English flu seasons, I have previously ran a granger test which has proved that the Australian figures seem to have similar ...
1
vote
1answer
29 views

Seasonal difference in ARIMA

I have time series with frequency=7. ndiffsfunction (https://www.rdocumentation.org/packages/forecast/versions/8.10/topics/ndiffs) suggests first order difference ....
0
votes
0answers
24 views

Large time series for short-term forecasting [closed]

Let’s assume I have a long time series of e.g. a stock price over the last 10 years with hourly data. Now I am interested in doing short-term predictions, i.e. to predict the price for the next few ...
0
votes
1answer
40 views

Simulate multiple forecasts with fixed accuracy

I have a time series forecast along with actual historical data, and its accuracy (MAPE, probability coverage etc.) is calculated. Now I want to estimate how improving some or all of the accuracy ...
0
votes
0answers
13 views

Examine the below time-series plot of data. With reference to the graph, justify the chosen method for the analysis that has been started on the table

Please can I be provided with a good answer/justification for the question? I think there is one method and it is moving average? I am not sure though.
0
votes
0answers
11 views

How to interpret periodogram of py script results

I have 5 years weekly sales data,did primary detrending(using seasonal decomposition) on it and processed the dtrended data through periodogram script(Welch code) and got below results. However,am ...
3
votes
1answer
41 views

Dynamic Time Warping Univariate Time Series to aid in selecting Forecasting Model

I have approximately 174 univariate time series that I would like to forecast. These are all country observations that have been thoroughly cleaned with no outliers or missing values. I would like to ...
0
votes
1answer
55 views

Based on the graph & table, what method is used for the analysis that has been started on the table pictured?

Looking at the time-series plot of data (pictured), and looking at the table (pictured), what method and why has been chosen for the analysis that has been started on the table shown? I'm struggling ...
0
votes
0answers
16 views

Comparing two tests: Diebold-Mariano vs. Giacomini-White

What is (are) the main difference(s) between the Diebold-Mariano and the Giacomini-White tests of superior predictive ability? When does (do) the difference(s) matter?
0
votes
0answers
18 views

Does the steps outlined are the right way of finalizing and saving ETS models to perform forecasting using python?

I have 5 years of weekly sales data of different retail products and trying to implement ETS modeling.Data split has been done in a way that last one year data goes into test and remaining in train ...
1
vote
1answer
41 views

Should I use the Diebold-Mariano test year-by-year or on the overall forecast?

I have built two models, one ARIMAX and one VAR, to compare against a baseline ARIMA model to predict a weekly economic time series of interest. I am primarily comparing the accuracy of my models ...
0
votes
0answers
20 views

How to perform multi-step forecasting in holt-winter modeling in python

I am trying to implement demand forecast of a produt for a month ahead and then convert it to week wise. I have received 7 years of week wise sales data,did all necesary data preprocessing,train(up to ...
2
votes
1answer
68 views

Forecasting with mixed models

I need to forecast sales for a data set where I have the amount sold per item and week. There are also categorical variables that are supposed to be integrated into the model as random effects. What ...
1
vote
0answers
18 views

Fitting a Local Poisson model (Exponential Smoothing) [closed]

I am working through "Forecasting with Exponential Smoothing". I am stuck on exercise 16.4 on the part that states: The data set partx contains a history of ...
0
votes
0answers
15 views

How does clustering will affect the MAPE and RMSE?

MAPE and RMSE are two very popular techniques to calculate the error. Now assume I have time series and cluster them to K clusters. This might reduce the training time when we are using the ...
1
vote
0answers
36 views

Have log returns series almost always conditional mean zero? I presume no

I'm analyzing S&P500 stocks daily log-returns on the 505 time series of the biggest companies in the USA between 2014-01-01 and 2019-12-01. My task was to identify the ARMA-GARCH model of them. ...
1
vote
0answers
16 views

Including the 'xreg' argument in the arfima() function of the 'forecast' package? [closed]

after successfully implementing the xreg argument in the arima() function, I would like to somehow include the external regressors into an ARFIMA model because my time series (realized volatility of ...
0
votes
0answers
14 views

Why do test set predictions perform far better than a recursive forecast - time series forecast

I've been dealing with a LSTM stock forecaster, and I've been looking at articles like 1, 2. I know that the models are very likely overfitted, but nonetheless, the test set predictions are quite ...
2
votes
1answer
19 views

Higher RMSE lower MAPE

I have a time series model that forecast next K days. For example when I forecast next 50 days my MAPE is 20.3% and RMSE is 2943 and when I forecast next 200 days is the MAPE is 10.25 % but RMSE is ...
2
votes
1answer
27 views

General exponential smoothing to linear functions of past observations

I am just trying to derive an equation in "Forecasting with Exponential Smoothing" page 36 section 3.2. I am given the following $\hat{y}_{t|t-1} = \textbf{w}'x_{t-1}$ $\epsilon_{t} = y_t - \hat{y}...
0
votes
0answers
17 views

How to make StatsModels ARIMA more accurate?

I'm working on a big data project for my school project. My dataset looks like this: https://www.kaggle.com/umar47/usd-try I've translated column names to English to work smoothly. New column names ...
1
vote
2answers
44 views

Forecasting using regression model

I am regressing gross sales against two regressors X1 and X2, and have a linear regression model with me. I want to use this model to get forecasts far out into the future where the values of X1 and ...
1
vote
0answers
16 views

Can you recursively forecast one series with two series?

My question is probably very elementary but I haven't been able to find an explanation of recursive forecasting that I fully understand. I've read a journal article that seemed to recursively ...
1
vote
1answer
37 views

How do we forecast using 3 point moving average?

X<- c(3,6,8,10,6,5) If I want to forecast using 3point moving average I use ma(X,3) from forecast package So this is going to give a series of smoothed average. If I want to forecast further 2 ...
0
votes
0answers
25 views

Forecasting with no seasonality

I have a transactions data frame and a promotions data frame. And I want to perform a forecast. The problem is that I can't tell if my data presents seasonality or not. I mean the sales per week looks ...
1
vote
1answer
48 views

Why do the regression residuals from a regression model with ARIMA errors differ from residuals from a linear regression model?

Let’s start loading fpp3 package (https://github.com/robjhyndman/fpp3-package) and the US consumption expenditure dataset (https://rdrr.io/cran/fpp3/man/us_change.html) ...
1
vote
1answer
30 views

Correlation between features in time-series

This is a technical/conceptual question. I am not sure if this is the right place to ask. If not, please let me know, I will change it. Question: I have some time series data with 12 room ...
1
vote
1answer
12 views

Scaling the MAE by the mean of non zero points for intermittent data

I am currently trying to find a way of scaling the MAE for my intermittent data. The data is always greater than 0 and is intermittent, with long periods of zeros. I have read a few papers that ...
1
vote
1answer
43 views

Recursive or direct forecasting used in forecast/predict() in stats models

I am working on a time series project. I have an hourly series and I have to forecast the 24 next hours. I am facing a problem with understanding how both the stats models forecast() and predict() ...
0
votes
1answer
43 views

R forecasting model for daily time series with a lot of zeros with annual seasonality

I have been trying to do the forecasting model. There is biological data (pollen grain in the air, which appears each year during the spring/summer time). My data has daily value and there is an ...
1
vote
1answer
41 views

How to know /guess which series(out of 100) will give better forecast before applying any time series model and without plotting each series?

I have More than 100 variables which are to be forecast. But before applying any time series model and plotting series, How can I guess that a particular series may have a higher chance of getting ...
1
vote
0answers
44 views

“Back-casting” a Time Series

The problem I have is that i have a series of data (between 2000-2010) and i have another series (independant variable) which is available between 1980 and 2010. I need to backcast the value of the ...
0
votes
1answer
18 views

RNN model for predicting room temperatures

I am currently doing a project in Machine Learning where I am trying to predict the temperature of a room in future. I have a 1-year dataset of a house with 12 rooms. Data is collected at 10 min ...
0
votes
1answer
6 views

What type of panel data model to run for forecasting on a site level basis?

I am trying to forecast monthly energy volume on a site-level basis (around 2,000 individual sites). I have monthly data for each site for at least two years. I also have attributes such as square ...
1
vote
0answers
22 views

Interval and density forecast in R with both heteroskedasticity and non-normality in time-series data

We tried to get both an interval and density forecast based on time-series data, which we found to be both non-normal and heteroskedastic, in R. We know that for non-normality, forecasts can be ...
0
votes
0answers
20 views

Use R forecast::arfima to forecast anti-persistent series - REPOST - more nicely formatted

I'm trying to use the R forecast package to forecast an anti-persistent time-series (assumed to be an ARFIMA(0, d, 0) series, with d somewhat negative, e.g. d = -0.25). The forecast::arfima function ...
0
votes
0answers
13 views

Use R forecast::arfima to forecast anti-persistent series [duplicate]

I'm trying to use the R forecast package to forecast an anti-persistent time-series (assumed to be an ARFIMA(0, d, 0) series, with d somewhat negative, e.g. d = -0.25). The forecast::arfima function ...
1
vote
0answers
19 views

Product Demand Forecasting for Mutliple Products in Single Warehouse

I am working on a new project I haven't much experience with and was looking for insight on where to begin and methods to use. I am trying to produce a demand forecasting model (or perhaps sales ...
1
vote
1answer
32 views

Time series when seasonality appear due to both solar and the lunar calendars

I have a time series data as shown in the figure below where the X axis is the serial number of the day of the year form 1 to 365 where 1 is 1-Jan and 365 or 366 is 31-Dec. The Y axis represents the ...
2
votes
1answer
35 views

Is it necessary to remove Seasonality while time series forecasting using ML methods ? Can't model learn it on itself?

I think ML model can learn from seasonal variations also. But if we remove seasonal variations, model & add it back, then essentially, we will end up dividing learning into : 'seasonal variations ...
0
votes
1answer
18 views

Forecasting a not-seasonal time series in R

I would to forecast a not-seasonal time serie in R. This is my serie and the model built with HoltWinters: ...